This article was based on the interview with Athos Commerce CMO Gary Lombardo on why only 14% of shoppers see one consistent brand by Greg Kihlström, AI and MarTech keynote speaker for The Agile Brand with Greg Kihlström podcast. Listen to the original episode here:
For those of us who have built our careers in marketing, the funnel has been a reassuringly solid structure. It was our map, our model, our source of comfort in the chaotic world of consumer behavior. We drew it on whiteboards, built our teams around its stages, and spent countless hours (and dollars) optimizing every transition from awareness to purchase. We knew the funnel was never a perfect representation of reality, but it was a useful fiction. Today, however, that fiction is becoming untenable. The map no longer reflects the territory, because a significant portion of the territory is now a black box.
When upwards of 60% of consumers begin their shopping journey with an AI tool, as recent research shows, they are not entering the top of your funnel. They are entering a new discovery engine that you don’t own, can’t instrument, and barely understand. This great fragmentation—across AI, social discovery, and marketplaces—isn’t just adding a few new channels to the marketing mix. It’s fundamentally breaking the models we’ve relied on for decades. This raises some uncomfortable but necessary questions for marketing leaders. If your website is no longer the definitive starting point, what is your most important asset? And how do you build a brand that behaves like one company when it shows up in countless places you don’t control? I recently explored this with Gary Lombardo, Chief Marketing Officer (CMO) at Athos Commerce, a company on the front lines of helping over 2,700 global brands navigate this new reality. His insights suggest the answers lie less in new campaign tactics and more in a radical focus on the data foundation of your entire operation.
When the Journey Starts Elsewhere, Your Org Chart Breaks First
The first casualty in this new world isn’t your attribution model or your campaign ROI; it’s something far more fundamental. When a customer’s discovery process moves from your well-lit website to the opaque world of a large language model, the first thing that cracks is often the marketing org chart itself. We’ve built our teams into efficient, channel-specific silos because it made sense. There’s a team for paid search, a team for the website, a team for email. They have clear KPIs, established playbooks, and direct control over their domain. But as Lombardo points out, this structure is ill-equipped for a world where the most critical touchpoints happen off-platform.
“Most large marketing teams are still structured around channels that they can own and measure, like paid, the website, owned email… The moment that 60% of your customers are researching through a tool you don’t own, don’t control, and can’t instrument with your usual analytics tags… you’ve got a structural gap. There’s no one whose job it is to own what AI says about this.”
This structural gap is where brand consistency goes to die. The website team ensures product information is perfect on the product detail page. But who ensures that same information is accurately represented in a Gemini or ChatGPT response? In most organizations, the answer is no one. This isn’t a failure of talent, but a failure of design. It’s a quiet crisis of accountability that directly leads to the dismal statistic from Athos’s research: only 14% of consumers find product information to be consistent across channels. When the traditional funnel model—and the organizational structure built around it—no longer holds, the very concept of a cohesive customer experience begins to fall apart.
The Inconsistency You See is a Data Problem in Content’s Clothing
When a leader sees an incorrect price on a marketplace or a missing size on a social post, the immediate instinct is to treat it as a content problem. We dispatch the channel manager or the content team to fix the offending asset, and the fire is temporarily put out. But this is treating the symptom, not the disease. The inconsistency that customers experience is rarely the fault of a single lazy entry; it’s the visible manifestation of a deeper, systemic issue. As Lombardo aptly puts it, what we are seeing is a data problem masquerading as a content problem.
“It’s tempting to call it a content problem because that’s the visible symptom… But if you trace it back, it’s almost always a data problem wearing content’s clothes, so to speak. Most brands don’t have one canonical structured source of truth for product. They have, you know, a PIM [Product Information Management system] that is 80% right, a feed file that gets patched maybe by hand by one channel, and a merchandiser spreadsheet that is more current than either.”
This patchwork of “mostly right” data sources is the silent killer of consistency. An 80% accurate PIM might have been acceptable when the primary output was your own website, where a merchandiser could manually override errors. But when that same data is being ingested by dozens of automated systems—from Google’s Merchant Center to an LLM’s training set—“mostly right” is functionally equivalent to “wrong.” The solution isn’t more vigilance from the content team; it’s a re-architecting of your information infrastructure. This shifts the ownership of the problem. If it’s a content issue, marketing owns it. But if it’s a foundational data problem, it becomes a shared mandate across marketing, e-commerce, IT, and operations, requiring a level of cross-functional collaboration that many organizations are not yet wired for.
From Retrieval to Synthesis: Why GEO Is Not the New SEO
With the rise of generative AI, the term “Generative Engine Optimization” (GEO) has entered the lexicon, and the natural reflex for many is to hand this new responsibility to the SEO team. After all, it sounds similar enough. This, however, is a critical miscalculation based on a misunderstanding of how these new systems work. We are not simply optimizing for a new kind of search engine; we are optimizing for a new kind of logic. Treating GEO as a simple extension of SEO is a recipe for frustration and wasted effort. Lombardo draws a clear and crucial distinction between the two disciplines.
“SEO is fundamentally a retrieval game. You’re optimizing to be one of the 10 blue links… that a human will scan and click. GEO is really a synthesis game. You’re optimizing to be the fact the model is choosing to restate. Oftentimes there’s no click at all… The mistake teams make is porting over the concept of SEO, specifically around keyword thinking.”
This is a profound shift. SEO is about ranking; GEO is about being understood. An SEO-driven approach might involve stuffing a product description with popular keywords. A GEO-driven approach involves meticulously structuring the unglamorous attributes of a product—dimensions, materials, compatibility, return policies, power consumption—so a machine can parse them as unambiguous facts. The goal is not to trick an algorithm into showing your link, but to provide such clear, comprehensive, and trustworthy data that the AI model chooses to use your product’s attributes as the basis for its synthesized answer. This means the work of GEO begins not with keyword research, but with the data hygiene we’ve already discussed. Your product feed is your new landing page.
In the Age of Agents, Precision Replaces Persuasion
Looking ahead, we are entering the era of what Athos calls “agentic commerce,” where a consumer’s AI assistant will conduct the initial product evaluation. This agent won’t be swayed by your beautiful hero image or your clever brand slogan. It will be comparing structured data points: price, availability, shipping speed, and a hundred other attributes. So, in a world where the first “customer” is a dispassionate machine, where does brand live? Lombardo argues that it doesn’t disappear, but its role changes. The key differentiator in this initial exchange shifts from emotional persuasion to factual precision.
“Differentiation shifts from persuasion to precision… Agents aren’t swayed by big beautiful hero images or clever headlines. It’s swayed by more accurate or complete structured attributes… Paradoxically, in the agentic world, being scrupulously honest and complete about your product data becomes a competitive advantage in its own right, and the brands that win are the ones an agent can trust enough to recommend confidently.”
This is perhaps the most significant strategic implication for marketing leaders. For a century, our primary currency has been persuasion. We build brands by telling compelling stories and forging emotional connections. That work remains vital—for the human who ultimately makes the purchase decision. But to even get to that human, we must first win a game of precision. As proof, Lombardo cites a department store client that saw a 66% performance uplift simply by tightening up how their product data was structured, with no creative changes. In this new landscape, brand building must expand to include the rigorous, detail-oriented work of data governance. Your brand’s trustworthiness is now a direct function of your data’s integrity.
The conclusion here is both daunting and empowering. The great fragmentation of product discovery requires more than just a new line item in the marketing budget for “AI initiatives.” It demands a fundamental rethinking of our internal structures, our data infrastructure, and our very definition of brand building. The silos that served us in the age of owned channels are now liabilities. The “good enough” data that powered our websites is now a source of brand-damaging inconsistency.
The work required is not the glamorous stuff of Super Bowl ads or viral campaigns. It’s the hard, cross-functional effort of creating a single, canonical source of product truth. It’s about elevating data governance from an IT concern to a core marketing competency. The leaders who embrace this shift—who understand that precision is the new persuasion and that their product data is now a primary expression of their brand—will be the ones who navigate this fragmented world successfully. They will build the brands that machines trust, and by extension, the brands that customers will continue to choose.
